Could pesticide toxicology studies be more relevant to occupational risk assessment?

Pesticide toxicology study design has evolved from concern for oral exposure via food residues. The emphasis on the oral route does not generally apply to workers that are exposed primarily via the dermal route either handling pesticides or re-entering treated fields. As a result numerous assumptions about how oral toxicology results relate to dermal exposure must be made when conducting worker risk assessments. These assumptions introduce a high degree of uncertainty. Alternative toxicology study designs are suggested to reduce uncertainty when assessing risk. Because the dermal route is so important to characterizing occupational risk, methods to improve the accuracy of dermal absorption estimates are suggested, including the use of human subjects to study dermal absorption. Additional suggestions include tailoring dermal, oral and inhalation kinetic study designs to reflect worker exposure dosages. Suggestions are made to routinely conduct a single dose toxicity study patterned after the neurotoxicity study design to distinguish single dose effects and NOAELs from those resulting from multiple doses. Finally, interspecies pharmacokinetics studies are proposed to determine which toxicology study regimen of dosing best reflects intermittent worker exposure.

Estimation of dermal absorption using the exponential saturation model.

Estimates of dermal absorption are used in exposure assessment to calculate the internal dose of persons contacting pesticides and are a critical part of risk assessments. An exponential saturation model with lag time was validated against a classic dermal absorption study of 12 pesticides administered to human volunteers. The model gave dermal absorption estimates consistent with reported values in the literature. Moreover, this model gave more realistic estimates of the percentage of dermal absorption for some pesticides, which have special properties. In most submitted dermal absorption studies in animals, especially rats, "bound" skin residues (BSR) at treated skin sites were generally high when animals were sacrificed more than 24 h after the dose was administered. The direct addition of the total BSR as an absorbed dose would likely overestimate actual dermal absorption. From a well-conducted dermal absorption study, this model can be utilized to estimate maximum excretion of the administered dermal dose as a result of further absorption of bioavailable BSR. Resulting dermal absorption estimates are appropriate for regulatory purposes in the risk assessment of pesticides because they take into account the bioavailability of BSR while at the same time the estimates are not overly conservative.

Metabolism and excretion of dimethoate following ingestion of overtolerance peas and a bolus dose.

Data to guide an exposure assessment were obtained by giving sugar peas containing overtolerance dimethoate residues (17 ppm; 8% oxon) and a bolus dose of dimethoate to a healthy adult male. The dimethoate tolerance on peas was and remains 2 ppm. Serial total urine samples were collected and analysed for dimethoate and its oxon, dimethylphosphate, dimethylphosphorothioate (DMTP) and dimethylphosphorodithioate. The dose of dimethoate administered was approx. 0.1 mg/kg body weight and produced no symptoms of toxicity. Dimethylphosphates appeared in the urine within 2 hr. The major metabolite (about 60%) was DMTP. Only traces (< 0.5%) of dimethoate and oxon were recovered from urine. Acetylcholinesterase inhibition was not observed although urinary metabolites were prominent, indicating that they are better indicators of acute exposure than cholinesterase inhibition. The results obtained using a bolus dose were virtually identical to those from the trial with overtolerance peas, and indicated that dimethoate is readily absorbed and its urinary metabolites are readily eliminated following exposures to low doses (0.1 mg/kg body weight).

Captan metabolism in humans yields two biomarkers, tetrahydrophthalimide (THPI) and thiazolidine-2-thione-4-carboxylic acid (TTCA) in urine.

Captan fungicide (N-(trichloromethylthio)-4-cyclohexene-1,2-dicarboximide) metabolism in two human volunteers rapidly yields THPI (tetrahydrophthalimide) and TTCA (thiazolidine-2-thione-4-carboxylic acid). The work was done to evaluate usefulness of TTCA and THPI as biomarkers of occupational exposure and to compare human and rat dermal absorption and metabolism. THPI in 12h urine ranged from MDL (5 ppb) to 640 ppb and was stable for at least one year. TTCA was also a stable metabolite, but the MDL was 50 ppb. THPI was detectable in urine for 72 hours following oral dosages of 1 mg/kg, but most was eliminated 0-24 h. No THPI was detectable in urine following application of a chloroform solution to hands, forearms, or inguinal region. Dermal absorption and metabolism of captan are substantially different in humans and rats.

Assessing human exposures to pesticides.

Pesticide use is inevitably associated with chemical exposures that range from inferred nondetectable levels to easily measurable ones using sensitive, readily available analytical tools. Whether these exposures are of any biological significance is determined by duration, dose, and biological reactivity. The overwhelming majority of human exposures occur in a diverse chemical milieu of a nutritive substances and are of no known significance. Technologies that minimize human chemical exposures and maximize pesticide effectiveness are favored. The risk characterization process is ideally suited to assist decision makers concerning the protection of human health and evaluation of agricultural tools. It is the best means available to balance the review of pesticide impacts on health and agriculture. Regulators must be cautious to acknowledge the relative rather than absolute nature of the risk characterization process. Workplace biological monitoring must become more commonplace as a means to evaluate the chemical exposure potential of various work tasks and greater attention must be given to the biological validation of methods. Earlier needs for data to develop workplace hygiene strategies have been replaced in recent years by demands of the risk assessment process, which utilizes direct estimates of exposure and absorbed dose. Animal models, no matter how attractive, are not presently a substitute for human experience. Opportunities to gather more information on human experience associated with pesticide exposures must be more aggressively identified and pursued. Only a very small time lag should exist between identification of pesticide metabolites in rats and evaluation of metabolic similarities in humans. At the present levels of analytical sensitivity, most of our current uncertainty about the extent of worker exposure and patterns of metabolism between species can be at least clarified with the cooperation of persons who are exposed during normal day-to-day activities in the workplace. Only with better human data will the risk assessment process warrant greater reliance in decision making concerning our chemical exposures and human experience.

Aflatoxin B1 hydroxylation: sensitive monooxygenase microassay suitable for liver biopsy specimens.

A method for measuring the rate of monooxygenase catalyzed hydroxylation of aflatoxin B1 to aflatoxin Q1 has been developed. It is suitable for use with small amounts of tissue obtained by needle or surgical biopsy of liver. Liver samples are homogenized in buffered saline containing a glucose-6-phosphate:NADPH generating system. 14C-AFB, is added to initiate the 10 min oxidation period. Quick-freezing stops oxidations, and AFB1 and AFQ1 are extracted using CHCl3. After solvent removal, the residue is resuspended in benzene-acetonitrile (98:2) and applied to silica gel thin-layer chromatography plates which are developed twice in chloroform-acetone-n-hexane (85:15:20). AFB1 and AFQ1 were located by fluorescence and scraped into vials for scintillation counting. The reaction was linear for 10 min using 5-10 mg liver tissue per 0.2 ml incubation. Aflatoxin M1 and 3 other metabolites more polar than AFB1 were also produced. The apparant Km for AFQ1 formation was 8.6x10(-4)M and Vmax was 287 picomoles AFQ1 mg protein-1 min-1 using bonnet monkey Macaca radiatea liver. The procedures should be valuable in comparative biochemical studies, especially those using liver biopsy specimens of primates.